rexresearch.com
Dong SEO,
et al.
Graphene 1-Step Bulk Production
[ EXCERPTS ]
https://www.nature.com/articles/ncomms14217
Nature Communications 8, Article number: 14217 (2017)
doi:10.1038/ncomms14217
Single-step ambient-air
synthesis of graphene from renewable precursors as
electrochemical genosensor
Dong Han Seo, et al.
Thermal chemical vapour deposition techniques for graphene
fabrication, while promising, are thus far limited by
resource-consuming and energy-intensive principles. In particular,
purified gases and extensive vacuum processing are necessary for
creating a highly controlled environment, isolated from ambient
air, to enable the growth of graphene films. Here we exploit the
ambient-air environment to enable the growth of graphene films,
without the need for compressed gases. A renewable natural
precursor, soybean oil, is transformed into continuous graphene
films, composed of single-to-few layers, in a single step. The
enabling parameters for controlled synthesis and tailored
properties of the graphene film are discussed, and a mechanism for
the ambient-air growth is proposed. Furthermore, the functionality
of the graphene is demonstrated through direct utilization as an
electrode to realize an effective electrochemical genosensor. Our
method is applicable to other types of renewable precursors and
may open a new avenue for low-cost synthesis of graphene films...
Here we present a single-step, rapid thermal synthesis of uniform
and continuous graphene films in an ambient-air environment, using
a cheap and renewable form of biomass, soybean oil, as the
precursor. To the best of our knowledge, this is the first time
that the synthesis of graphene film has been demonstrated in an
ambient-air environment without any compressed gases. Graphene
derived from this unique ambient-air process exhibits good and
tuneable film properties, which are comparable to those of
graphene synthesized with conventional methods 2,11. This
ambient-air process for graphene fabrication is fast, simple,
safe, potentially scalable and integration-friendly. Importantly,
it offers the scope to potentially address the critical roadblocks
towards large-scale, efficient graphene manufacturing.
Controlled synthesis of graphene in ambient-air environment
Currently, graphene synthesis involves several key factors need to
be improved: (i) lengthy high-temperature annealing processes to
increase the grain size of the metal catalyst used to form
graphene; (ii) utilization of purified and compressed gases to
offer a homogenous and controlled delivery of carbon source
materials; and (iii) the use of lengthy vacuum operation to avoid
the presence of any detrimental reactive oxygen species from air
2,4. To overcome these problems, we have designed a thermal CVD
process to produce graphene in an ambient-air environment that is
completely free of compressed or purified gases and requires
minimum processing time.
The process is schematically illustrated in Fig. 1a, in
which the precursor for graphene growth and a metal catalyst (for
example, Ni foil) are placed close together inside the heating
zone of a furnace, before heating the quartz tube. The quartz tube
is then sealed and the temperature is increased. During the
ramping stage, air inside the quartz tube is released through a
valve to maintain atmospheric pressure. Once the annealing stage
is complete, the sample is removed from the heating zone for rapid
cooling. Raman spectra of the samples grown at 800°C in the
ambient-air process indicated the presence of single-to-few layer
graphene films covering the surface of the growth substrate
(Fig. 1b).
Fig1b
In the standard operation, the catalyst is low-cost
polycrystalline Ni foil. Graphene growth occurs by thermal
reforming of a natural precursor, soybean oil, in a closed
ambient-air environment. Unlike conventional CVD methods or
conventional natural precursor methods for growing graphene, the
technique does not require any purified gases 8,9. Moreover,
expensive vacuum processing is avoided. The natural precursors
substituted for purified gases are cheaper and safer. By
restricting the air flow into the quartz tube, the transformation
of solid-state carbon into carbon dioxide or other gaseous species
is prevented. By controlling the temperature, cooling rate and
precursor amount, the process enables the growth of homogenous
graphene films of good quality. A comparison of the method with
other CVD processes is provided in Supplementary Tables 1 and
2.
The parameters observed to control the quality of graphene include
temperature, processing time, precursor, substrate and the
ambient-air environment. Nickel acted as a good catalyst for the
breakdown of precursor material (in this case, the soybean-oil
molecules) into smaller building units that are essential for the
synthesis of graphene 12.
To investigate how the transformation occurred in the process, we
have analysed the chemical composition of the annealed soybean
oils at different temperatures (Supplementary Fig. 1).
During the early stages of the annealing process, for instance at
300°C, the long carbon chains in the soybean oil precursor were
thermally dissociated into gaseous carbon building units such as
methyl and ethyl species (Supplementary Fig. 1a). Other
gaseous species were also generated, including hydrogen, water,
hydroxyls and carbon dioxide, as confirmed by mass spectrometry
(Supplementary Fig. 1b and c). Traces of heavier
hydrocarbons such as propane were also observed. Most of the oil
was vapourized by about 425°C and a rapid mass reduction of the
oil was observed by thermogravimetric analysis below 500°C
(Supplementary Fig. 1d). These building units present in the
vapour can diffuse through the tube during the heating stage. As
the temperature gradually increases to 800°C, these carbon
building units begin to dissociate into carbon atoms and dissolve
into the Ni bulk. The sample was annealed for 3?min at 800°C to
promote dissolution of carbon atoms in the Ni substrate. Finally,
following the rapid cooling stage, carbon segregates from the bulk
and crystallizes on the Ni surface forming graphene 12,13.
At elevated temperatures, long hydrocarbons in the oil decompose
in the presence of O2 to form water vapour. In particular, water
vapour can promote the etching of amorphous carbon deposits on the
Ni surface 14. As such, we did not observe the formation of
amorphous carbons in our sample. This also helps maintain the
catalytic activity of the Ni surface in breaking down the
precursor material 15. Moreover, we have conducted a detailed
analysis on the consumption of oxygen in the reactor during the
growth process (Supplementary Note 1). We found that the
precursor amount was critical for the consumption of reactive
oxygen species. In the optimal growth condition, a slight carbon
excessive environment is used to promote the growth of graphene
and deter the formation of amorphous carbon. On the other hand, an
over-excessive amount of precursor material led to an
oversaturation of deposited carbon in the bulk of Ni, and
subsequently, the crystallization of graphite on the Ni surface.
This may explain the resulting formation of thick graphene sheets
as observed in Supplementary Fig. 2a. Moreover, in the
case of an insufficient amount of precursor, oxygen species can be
present in the as-grown product in the form of C–O amorphous
carbons (Supplementary Fig. 2b), consistent with the
aforementioned calculations of oxygen consumption
(Supplementary Note 1) 16. These experiments indicate the
critical role of the thermally dissociated precursor materials
(that is, hydrocarbons) in consuming the reactive oxygen species
present in the ambient-air environment, which has a profound
effect in controlling the quality of the as-grown graphene films.
We have also noticed that a slow cooling can promote excessive
carbon segregation from the Ni bulk, which may account for the
observed formation of a graphite-like film (Supplementary Fig.
2c). Another parameter that significantly influences the
growth of graphene in the ambient air environment is the annealing
temperature. At an annealing temperature of 500°C, an incomplete
formation of the graphene film was observed (Supplementary Fig.
2d). This may be attributed to an insufficient amount of
energy to dissociate and reform the precursor material (that is,
hydrocarbon species) required for graphene formation. Conversely,
at a higher annealing temperature of 900°C, thicker graphene
sheets were observed (Supplementary Fig. 2e). This may
arise from the increased rate of carbon diffusion, segregation and
graphitization as a result of the elevated temperature.
Importantly, these parameters allow us to obtain graphene films
with tuneable average thickness and optical transmission, as
characterized by Raman spectroscopy and optical transmission (Supplementary
Fig. 3).
It is worth mentioning that graphene did not form on other growth
substrate materials with significantly lower carbon solubility
than Ni, such as the commonly used Cu foil. Moreover, we did not
observe graphene formation on graphitic surfaces such as woven
carbon cloth (Supplementary Fig. 4). This suggests that
the use of Ni (through, for example, carbon solubility, carbon
segregation ability, catalytic effect, possibility of formation of
oxide in air) and its interaction with the precursor material play
a critical role in enabling the growth of graphene films. We also
investigated the possibility of transforming other types of
renewable oil groups. In particular, we were able to demonstrate
the ambient-air growth of similar graphene films from other types
of triglyceride (carbon)-containing precursors such as butter
(Supplementary Fig. 5). As such, this method is versatile
and may be tailored to transform other renewable carbon-containing
natural precursors into graphene films...
Proposed mechanism of graphene growth in ambient-air process
The growth of graphene in an ambient-air environment may initially
seem counter-intuitive, as graphene is expected to be destroyed in
air at elevated temperatures (above 500°C). However, we
hypothesize that the unique processing conditions promote the
controlled synthesis of graphene films in an otherwise destructive
environment. Specifically, the thermally dissociated precursor
material decomposes in the presence of reactive oxygen species
from the ambient-air, leading to the formation of water vapour as
a by-product (Supplementary Fig. 1). The water vapour may help
suppress the deposition of amorphous carbon, promote the thinning
of graphene layers and maintain the catalytic ability of the Ni
substrate in breaking down the precursor material into smaller
building units necessary for the growth of graphene films.
To better understand the growth process and the possible
interaction with Ni substrate, we conducted experiments to probe
the surface composition of Ni foils following treatments at
elevated temperatures. In particular, we investigated the
composition of:
Ni foil heat treated in ambient environment without soybean oil,
where surface oxidation will be prevalent (Supplementary Fig.
8);
Ni foil heat treated in ambient environment with soybean oil,
following procedure as outlined previously for the growth of
graphene, where surface oxidation may be prevented (Supplementary
Fig. 9).
Our XPS analyses showed that when the Ni foil was heated in the
ambient environment without soybean oil, oxygen was easily
identified on the surface (Ni:O ratio of 1:1.83). However, when
the Ni foil was heated with soybean oil, the oxygen content was
significantly reduced (Ni:O ratio of 2.69:1). These results
indicated that the breakdown of soybean oil in the reaction
chamber provided a reaction pathway for the consumption of O2,
which consequently limited the surface oxidation of Ni at elevated
temperatures.
Thus, we propose a growth mechanism based on these supporting
evidences. First, soybean oil thermally dissociates into a range
of carbon building units, for example, CH3, C2H2 and other
species, at the ramping stage (Supplementary Fig. 1).
During this stage, molecular fragments of the precursor material
may react with and consume O2 inside the reaction chamber through
possible reaction routes as outlined in Supplementary Note 1.
Water vapour produced as a by-product of the consumption of O2 may
also help suppress the formation of amorphous carbon. The
formation of water was supported by the observation of water
condensation at the cool ends of the quartz tube outside the
heating zone. These molecular fragments may further decompose at
higher temperatures to provide a source of carbon dissolved into
the Ni foil. This is supported by the detection of an extended
nickel carbide peak in the XPS spectra of an etched graphene/Ni
sample (Supplementary Fig. 9). Then, growth of graphene can occur
through a combination of surface-mediated growth on the Ni foil
and precipitation from dissolved species when the sample is
cooled. The precipitation step is critical as we observed that the
cooling rate was important to control the thickness of the
graphene films (Supplementary Fig. 3)...
Graphene films demonstrate excellent functional properties and are
promising for diverse applications. However, the high cost and
complexities associated with graphene production impede its
commercial viability. To this end, we present a novel method for
the synthesis of graphene films, in an atmospheric-pressure,
compressed-gas-free ambient-air environment utilizing safe,
low-cost renewable precursors. This ambient-air method offers
numerous advantages over conventional thermal CVD techniques for
graphene synthesis, which critically rely on resource- and
time-consuming procedures (Supplementary Tables 1 and 2 and
Supplementary Note 3). Graphene films with good structural
and optoelectronic properties were obtained. On average, the
graphene film demonstrated an optical transmission of ~93.9%, a
sheet resistance of ~324Osq-1, Raman ID/IG ratio of 0.15–0.25 and
I2D/IG ratio of 0.95–1.50 and domain sizes ranging 200–500nm. We
exemplify the essential process parameters (for example, cooling
rate, precursor content, temperature and so on) to enable
controlled synthesis and tailored properties of the graphene film
in the ambient-air process. Further, we propose a mechanism for
the growth of graphene in the ambient-air process, based on depth
profiling of the as-grown film, analyses of the ambient-air
composition in the reaction chamber and reaction pathways for
precursor reforming into graphene. The functionality of the
graphene films was demonstrated through its direct integration as
an electrochemical genosensor, in which sensitive and selective
bio-detection was realized. Importantly, the ambient-air synthesis
of graphene films from renewable precursors offers numerous
advantages and opportunities for future streamlined integration
into large-scale production infrastructures and the realization of
diverse graphene-enabled technologies.
Ambient-air thermal synthesis of grapheme
The growth of graphene was carried out in a thermal CVD furnace
(OTF-1200X-UL, MTI Corp) with a quartz tube (100 cm in length, 5
cm in diameter). Polycrystalline Ni foils (25µm, 99.5%, Alfa
Aesar) were used as the growth substrate. The experimental
schematic is shown in Fig. 1. Briefly, two alumina plates were
placed in the heating zone of the furnace. One alumina plate was
loaded with 0.14 ml of soybean oil precursor and the other was
loaded with the Ni foil growth substrate. The openings of the
quartz tube were then sealed. The growth of graphene proceeds with
a gradual heating and fast quenching temperature profile. First,
the furnace temperature was raised to 800°C at a rate of 30°C
min-1. This was followed by holding at 800°C for 3 min. After the
growth step, the sample was immediately removed from the heating
zone to enable a rapid cooling (at approximately 25°C min-1) to
segregate the homogeneous and continuous graphene films. Owing to
the evaporation and thermal expansion of the precursor material, a
small build-up in pressure within the tube was observed.
Throughout the heating stage (200 to 800°C), atmospheric pressure
was maintained in the quartz tube by allowing this build-up of
gases to exit via the exhaust of the tube. A controlled gas
environment was created in the tube through enabling the
circulation of gases produced by precursor evaporation. Following
the heating stage, pressure within the quartz tube was observed to
be stabilized at atmospheric pressure. No additional gases were
introduced into the quartz tube throughout the entire growth
process.
Transfer of grapheme
A poly (methyl methacrylate) (PMMA)-assisted transfer of graphene
was adopted. Briefly, 46mg?ml-1 of PMMA (Mw=996,000;
Sigma-Aldrich) was spin-coated onto the as-grown graphene on Ni
foil (3,000 rpm. for 1 min). The sample was then dried in open air
for 12?h. Subsequently, the underlying Ni foil was dissolved in 1M
FeCl3 in 30 min. The PMMA/graphene film then floated to the
surface. This was washed several times with deionized (DI) water.
Next, the PMMA/graphene was lifted off from the DI water bath and
transferred onto a glass substrate. The PMMA was then dissolved
with acetone, and the sample was repeatedly washed with DI water.
The graphene on glass was then used for subsequent microscopy and
electrical characterization...
References
1. Kong, J., Soh, H. T., Cassell, A. M., Quate, C. F. & Dai,
H. Synthesis of individual single-walled carbon nanotubes on
patterned silicon wafers. Nature 395, 878–881 (1998).
2. Bae, S. et al. Roll-to-roll production of 30-inch graphene
films for transparent electrodes. Nat. Nanotechnol. 5, 574–578
(2010).
3. Yang, F. et al. Chirality-specific growth of single-walled
carbon nanotubes on solid alloy catalysts. Nature 510, 522–524
(2014).
4. Kim, K. S. et al. Large-scale pattern growth of graphene films
for stretchable transparent electrodes. Nature 457, 706–710
(2009).
5. Novoselov, K. S. et al. A roadmap for graphene. Nature 490,
192–200 (2012).
6. Ren, W. & Cheng, H. M. The global growth of graphene. Nat.
Nanotechnol. 9, 726–730 (2014).
7. Zurutuza, A. & Marinelli, C. Challenges and opportunities
in graphene commercialization. Nat. Nanotechnol. 9, 730–734
(2014).
8. Sun, Z. et al. Growth of graphene from solid carbon sources.
Nature 468, 549–552 (2010).
9. Ruan, G., Sun, Z., Peng, Z. & Tour, J. M. Growth of
graphene from food, insects, and waste. ACS Nano 5, 7601–7607
(2011).
10. Guermoune, A. et al. Chemical vapor deposition synthesis of
graphene on copper with methanol, ethanol, and propanol
precursors. Carbon 49, 4204–4210 (2011).
11. Reina, A. et al. Large area, few-layer graphene films on
arbitrary substrates by chemical vapor deposition. Nano Lett. 9,
30–35 (2009).
12. Li, X., Cai, W., Colombo, L. & Ruoff, R. S. Evolution of
graphene growth on Ni and Cu by carbon isotope labeling. Nano
Lett. 9, 4268–4272 (2009).
13. Yoon, S. M. et al. Synthesis of multilayer graphene balls by
carbon segregation from nickel nanoparticles. ACS Nano 6,
6803–6811 (2012).
14. Hata, K. et al. Water-assisted highly efficient synthesis of
impurity-free single-walled carbon nanotubes. Science 306,
1362–1364 (2004).
15. Zhu, L., Xiu, Y., Hess, D. W. & Wong, C. P. Aligned carbon
nanotube stacks by water-assisted selective etching. Nano Lett. 5,
2641–2645 (2005).
16.Ferrari, A. C. et al. Raman spectrum of graphene and graphene
layers. Phys. Rev. Lett. 97, 187401 (2006).
17.Garlow, J. A. et al. Large-area growth of turbostratic graphene
on Ni(111) via physical vapor deposition. Sci. Rep. 6, 19804
(2016).
18. Niyogi, S. et al. Spectroscopy of covalently functionalized
graphene. Nano Lett. 10, 4061 (2011).
19. Tuinstra, F. & Koenig, J. L. Raman spectrum of graphite.
J. Chem. Phys. 53, 1126–1130 (1970).
20. Hwang, J. Y., Kuo, C. C., Chen, L. C. & Chen, K. H.
Correlating defect density with carrier mobility in large-scaled
graphene films: Raman spectral signatures for the estimation of
defect density. Nanotechnology 21, 465705 (2010).
21. Nair, R. R. et al. Fine structure constant defines visual
transparency of graphene. Science 320, 1308–1308 (2008).
22. Yick, S., Han, Z. J. & Ostrikov, K. Atmospheric
microplasma-functionalized 3D microfluidic strips within dense
carbon nanotube arrays confine Au nanodots for SERS sensing. Chem.
Commun. 49, 2861–2863 (2013).
23. Marcano, D. C. et al. Improved synthesis of graphene oxide.
ACS Nano 4, 4806–4814 (2010).
24. Turner, A. P. F. Biosensors: sense and sensibility. Chem. Soc.
Rev. 42, 3184–3196 (2013).
25. Ferri, C. P. et al. Global prevalence of dementia: a Delphi
consensus study. Lancet 366, 2112–2117.
26. Cheng, L. et al. Prognostic serum miRNA biomarkers associated
with Alzheimer/'s disease shows concordance with
neuropsychological and neuroimaging assessment. Mol. Psychiatry
20, 1188–1196 (2015).
27. De Jager, P. L. et al. Alzheimer's disease: early alterations
in brain DNA methylation at ANK1, BIN1, RHBDF2 and other loci.
Nat. Neurosci. 17, 1156–1163 (2014).
28. Lunnon, K. et al. Methylomic profiling implicates cortical
deregulation of ANK1 in Alzheimer's disease. Nat. Neurosci. 17,
1164–1170 (2014).
29.Suni, I. I. Impedance methods for electrochemical sensors using
nanomaterials. Trends Anal. Chem 27, 604–611 (2008).
30. Hu, Y. et al. Decorated graphene sheets for label-free DNA
impedance biosensing. Biomaterials 33, 1097–1106 (2012).
31. Zhang, Z. et al. Tryptamine functionalized reduced graphene
oxide for label-free DNA impedimetric biosensing. Biosens.
Bioelectron. 60, 161–166 (2014).
32. Zhang, J. et al. Scaly graphene oxide/graphite fiber hybrid
electrodes for DNA biosensors. Adv. Mater. Interfaces 2, 1–6
(2015).
33. Armbruster, D. A. & Pry, T. Limit of blank, limit of
detection and limit of quantitation. Clin. Biochem. Rev 29,
S49–S52 (2008).
Dong Han Seo
WO2017027908
GRAPHENE SYNTHESIS
Inventor(s): SEO DONG HAN, et al.
The invention relates to methods for the production of high
quality graphene. In particular, the invention relates to
single-step thermal methods which can be carried out in an
ambient-air or vacuum environment using renewable biomass as a
carbon source. Specifically, the invention comprises heating a
metal substrate and carbon source in a sealed ambient environment
to a temperature which produces carbon vapour from the carbon
source such that the vapour comes into contact with the metal
substrate, maintaining the temperature for a time sufficient to
form a graphene lattice and then cooling the substrate at a
controlled rate to form a deposited graphene.
The invention relates to methods for the production of high
quality graphene. In particular, the invention relates to
single-step thermal methods which can be carried out in an
ambient-air or vacuum environment using renewable biomass as a
carbon source.
Background Art
Graphene exhibits unique electronic, optical, chemical and
mechanical properties. Because of its extremely high electron
mobility (electrons move through graphene about 100 times faster
than silicon), very low absorption in the visible spectrum and
relative flexibility and elasticity (compared to inorganics such
as indium tin oxide), supported horizontal graphene as an active
functional material has been revolutionising many fields. For
instance, graphene is potentially useful for flexible,
transparent, and wearable electronics, in energy storage devices
(e.g., fuel ceils, supercapacitors, photovoltaics, lithium-ion
batteries, etc), in devices for diagnostics and therapeutics
(e.g., biosensors, bioelectronics, drug delivery), in water
purification (e.g., point-of-use filtration membranes) and in
catalysis (e.g., to promote hydrogen evolution reactions). Control
of defect content, micro-structure, and surface chemical
properties in the graphene will be critical to maximising the
potential of graphene in these applications.
Graphene can be produced by a variety of methods. The mass
production of graphene, which would be essential for widespread
commercial use, has to date been targeted by a small number of
general processes, most notably:
mechanical grinding of graphite and dispersion in solution
followed by self assembly.
thermal graphitisation of SiC.
chemical vapour deposition (CVD) onto metal substrates.
Of these three methods, CVD onto metal substrates is the most
promising, as it produces graphene films of sufficiently high
quality to allow the potential of graphene to be more fully
realised. CVD also allows roli-to-roli graphene synthesis. The
quality of the graphene produced is critical to its ability to
function as a high performance material. High quality graphene
possesses a minimal number of defects from the ideal perfectly
regular sp<2>carbon film, and is also very thin, that is,
the bulk material produced contains as few carbon atomic layers as
possible. The qualify of graphene can be expressed quantitatively
in terms of its electronic and optical performance. A low number
of defects leads to a very low film resistance, which can
typically be around 200 Q/sq. Defects in the graphene can diminish
in-plane charge carrier transport which compromises the promising
properties required for efficient field-emission, ultra-fast
sensing and nano-electronics based devices. Very thin films, for
instance those having only one, two or three carbon atomic layers
are highly transparent and have a transmittance of up to 97% which
is useful for optical displays.
Thicker films and graphene in other forms (such as grains and
coatings) can be useful in other circumstances, such as catalysis
and filtration. An ability to control the thickness of graphene
grown is highly desirable.
However, CVD onto metal substrates has some inherent limitations.
The CVD apparatus itself is complex and expensive. CVD consumes
very large amounts of power and like other thermal methods
currently used, requires a low-pressure vacuum environment. This
means that there are significant capital and ongoing operating
costs associated with CVD. Also, the cost of vacuum equipment
increases exponentially with the size of the vacuum chamber which
limits the manufacturers' ability to scale up the process in a
cost effective manner.
CVD also requires the use of highly purified feedstock gases,
which are expensive. The use of gases such as hydrogen for
substrate passivation and methane and ethylene as carbon source
gases also means that additional hazard protection also needs to
be put in place.
CVD also requires relatively long time frames, of the order of
hours , for the growth, annealing and cooling steps to take place.
This inherent requirement means that CVD is not readily amenable
to the rapid mass production of affordable graphene.
The search for new methods of graphene is a very active area of
endeavour and many researchers are investigating synthetic routes
to high quality graphene thai are safe, inexpensive and amenable
to scale up.
For instance, US847Q400 discloses graphene synthesis from chemical
vapour deposition that take place in low pressure environments.
However, like conventional methods, the methods disclosed in US
8470400 require the use of purified gases, high temperatures and
long processing times. WO2013036272 discloses a crystalline
graphene and a method for the preparation thereof. This method
involves very high temperatures and long processing times. It uses
multiple processing stages and requires harsh chemicals and
processing environments.
US 81 /829423 discloses the combustion synthesis of graphene and
carbon nanomateriais. The method operates at very high
temperatures and high pressures and requires long processing
times. WO2013066269 discloses the synthesis of carbon
nanostructures from liquid carbon precursors using chemical vapour
deposition. However, the methods disclosed in WO2013066269 require
the use of purified gases, multiple processing steps, and long
processing times, This snapshot of a handful of current cases
illustrate the fact thai "state-of-the-art" graphene syntheses
require high temperatures, long processing times and in may cases
expensive apparatus and consumables.
It is an object of the present invention to overcome or ameliorate
at least one of the disadvantages of the prior art, or to provide
a useful alternative.
Any discussion of the prior art throughout the specification
should in no way be considered as an admission thai such prior art
is widely known or forms part of common general knowledge in the
field.
Summary in a broad aspect, the invention provides a method of
preparing a deposited graphene comprising the steps of heating a
metal substrate and carbon source in a sealed ambient environment
to a temperature which produces carbon vapour from the carbon
source such that the vapour comes into contact with the metal
substrate, maintaining the temperature for a time sufficient to
form a graphene lattice and then cooling the substrate at a
controlled rate to form a deposited graphene. The deposited
graphene is for preference in the form of a film, most preferably
a continuous film. For instance, the invention provides a method
of preparing a graphene film comprising the step of heating a
metal substrate (e.g., nickel or copper) and carbon source in a
sealed ambient environment to a temperature which produces carbon
vapour from the carbon source and dissolves carbon into the metal
substrate, maintaining the temperature for a time sufficient to
form a graphene lattice and then cooling the substrate at a
controlled rate to segregate the graphene lattice from the metal
substrate to thereby form a film of graphene.
Unless the context clearly requires otherwise, throughout the
description and the claims, the words "comprise", "comprising",
and the like are to be construed in an inclusive sense as opposed
to an exclusive or exhaustive sense; that is to say, in the sense
of "including, but not limited to".
As used herein, the terms "segregate", "segregates", "segregation"
and the like refer to the process whereby carbon atoms that were
diffused inside the metal substrate lattice precipitate out onto
the metal substrate surface. The metal substrate may be a
transition metai substrate, most advantageously nickel or copper.
The Ni metal substrate is preferred at a purity of 99% and above,
if Nickel is used, it is preferably in poiycrystalline form.
The metal substrate can be in any suitable form. Most usually it
will be in the form of a film but it can also be in the form of
grains, particles or even manufactured articles. When the metai
substrate is in the form of particles, it is to be understood that
the graphene exists discretely on the surface of the particles,
for instance, a metal particle coated with one of more continuous
layers of graphene.
The carbon source may be advantageously be biomass or derived from
biomass or purified biomass. The biomass or purified biomass may
be for example a long chain triglyceride (fatty acid), such as
soybean oil, or it may be a ceiiuiosic material. Renewable biomass
may be used. The carbon source may be in any form, such as liquid
or solid form with liquid usually advantageous.
The ratio of carbon source to metai substrate area is desirably in
the range of 0.1 to 0.3g of carbon source for each 10cm~ of
substrate, that is, 0.01 -0.03 mL/cm<2>. For thin films,
0.01 -0.025 ml/cm<2>is suitable, whereas for thicker films
0.025 - 0.03 ml/cm<2>or more can be used.
The metal substrate and carbon source are placed adjacent each
other, so that the metai substrate and the carbon source are both
located in the one heating zone. The substrate and source are
placed such that vaporized precursor is able to contact the
substrate. The ambient environment may be air at atmospheric
pressure or a vacuum.
In one embodiment, the ambient environment is air at atmospheric
pressure. Although the present invention is described with
reference to air, artificially prepared gases or combinations of
gas that mimic the action of air could be used if desired. Such
artificial combinations of gases could be used at pressures to
mimic the effect achieved by air at ambient pressure.
In another embodiment, the ambient environment is a vacuum,
preferably less than 1 mm Hg. in particular embodiments, it is
preferred that the metal substrate is nickel and the ambient
environment is air at atmospheric pressure. Alternatively, if is
preferred that the metal substrate is copper and the ambient
environment is a vacuum.
Unlike most methods in the art, the methods of the present
invention are free from the use of a compressed gas or gases.
Feedstock gases are not required. In one embodiment the sealed
environment is contained in any inert container, for example a
quartz or other dielectric heat resistant container such as a
quartz, alumina or zirconia tube.
The temperature sufficient to form a graphene lattice is
preferably in the range 650°C - 90Q<!>>C. The temperature
sufficient to form a graphene lattice is preferably maintained for
3 to 15 minutes depending on the purity of the metal foil.
Maintaining the temperature at around 800°C, for a time period
between 3-15 minutes result in upto about 19 layers of graphene. A
shorter time frame can be used to making a single layer of
graphene. An increased time period can be used to form graphen
films of 1 -3 layers in thickness, increasing the time period for
upto 15 mintutes can be used to make graphene upto 19 layers in
thickness.
If a thicker graphene film is required a higher growth temperature
is preferred, for example, maintaining the temperature 875°C or
above, or even around 900°C for a time period between 3- 5 minutes
result in greater than 20 layers of graphene. Maintaining the
temperature for a shorter period of time at 900°C is preferred for
forming graphenes of 20-40 layers. Increasing the time period upto
15 mintues result in increased layers of graphene of 40 layers or
more (>40 layers).
Once graphene reaches around about 40 layers or more, it may be
regarded as a form of ultra-thin graphite. However, it will be
understood that while the methods of the present invention refer
to graphene, they also encompass any form of layered or
multilayered continuous sp2 carbon sheets, such as ultrathin
graphite, that can be prepared by the method of the present
invention.
Preferably the substrate is cooled to ambient temperature at a
controlled rate of 10-100°C/minute. Cooling the substrate at a
controlled rate may be, for example, cooling the substrate back to
ambient temperature at a rate of 50-100°C/minute, or
alternatively, the substrate may be cooled at a rate of
25-50°C/minute, or 25-40°C/minute, or 15-25°C/minute or
5-15°C/minute or less than 5°C/minute. Examples include cooling
the substrate at a rate of up to 18°C/minute or at a rate of up to
25<!>>C/minute. The method may also further comprise the
step of decoupling graphene from the substrate.
As used herein, the term "decouple" "decouples", "decoupling" and
the like refer to the removal or lifting of a formed graphene from
the underlying substrate to isolate a graphene film. The graphene
may be decoupled by any conventional means. For instance, the
graphene may be decoupled from the underlying metal substrate by
dissolving the substrate in an acidic environment, in particular,
a nickel substrate may advantageously be dissolved in H2S0 or HCI
or FeCi3or a copper substrate may be dissolved in any of the
preceding or HN03. The invention also relates to as-grown and
free-standing graphene films prepared by the methods of the
invention.
Brief Description of the Drawings
Figure 1 shows a description of a preferred embodiment of
the process of the present invention.
Figure 2 shows (a) the proposed mechanism for graphene
growth, and (b) the temperature profile of the growth technique
of the present invention.
Figure 3 shows structural characterizations of graphene of
the present invention including, (a) optical micrograph, (b)
low-resolution TEM with electron diffraction pattern (inset),
(c) FFT of graphene, (d)-(e) HRTE showing 1 -3 graphene layers,
respectively.
Figure 4 shows Raman spectroscopic characterisations s for
the transformation of soybean oil precursor with increasing
annealing temperatures at (a) 500 °C, (b) 650 °C, and (c) 800
°C.
Figure 5 shows miscellaneous properties of the graphene of
the present invention, including (a) XPS survey scan, (b) C I s
scan, (c) transmittance spectra and (d) conductivity
measurements.
Figure 6 shows (a) the use of graphene of the present
invention in a biosensor assembly and (b)~(c) respective
biosensing results.
Figure 7 shows control of the thickness and quality of
graphene films wiih change of growth conditions, as
characterized by optical transmission and Raman spectral
mapping.
Figure 8 shows graphene growth at 9Q0°C as characterized by
Raman spectral mapping.
Figure 9 shows Raman spectral characterization of graphene
grown on copper foil.
Description
With reference to Figure 1 , the method of the present invention
is carried out in a sealed container (1) in an oven.
Typically, the container (1) is an inert tube, for example a tube
made from quartz, alumina, zsrconia or similar. The size of the
container is chosen so as to be relatively compatible with the
substrate being coated, that is, it is desirable to minimize the
amount of dead space in the container. The oven can be any type of
oven suitable for heating the container to temperaiures of the
order of 800°C. One type of suitable oven was found to be a
thermal CVD furnace (OTF-1200X-UL, MTi Corp), which is adapted to
heat tubular vessels. One example of a suitable tubular vessel is
a quartz tube of 100 cm length and 5cm diameter.
The method of the present invention involves placing a growth
substrate (2) and carbon source (3) in relatively close proximity
to one another in the container. They may be placed directly into
the tube, or more usually, are placed in inert crucibles (4), such
as alumina crucibles, prior to placement in the tube. The
container is then sealed and placed in the oven, or alternatively
placed in the oven and sealed. When the metal is Nickel, no gas
evacuation or flushing is required and the atmosphere in the
sealed container at the commencement of the process is air. An
ordinary mechanical seal will suffice. There is no need for the
container to be sealed to withstand significant pressure
differences.
The metal substrate and carbon source are placed adjacent each
other. The exact distance is not critical, as long as both the
substrate and carbon source are within the heating zone. Due to
the rapid thermal expansion of the vapours from the carbon source,
the concentration of vapours will be fairly consistent across the
heating zone. A degree of vacuum can be applied to aid in the flow
of precursors within the heating zone if required.
Normally, as the distance between substrate and source increases,
a larger amount of carbon source is required to cover a given
area. The exact parameters can be readily determined
experimentally for a given substrate size, carbon source and
target graphene quality.
The positioning of the carbon source and substrate within the
container should be such that when the container is in the oven,
the carbon source and substrate are both simultaneously within the
heating zone (5).
The substrate is a metal substrate, most desirably a transition
metal substrate, for example a nickel substrate, it has been
established by the inventors that there is little advantage to be
gained from using nickel that is higher than 99.5% purity. 99.9%
pure nickel or higher are suitable for use in the present
invention, but they produce no discernible advantage over 99.5% or
99% pure nickel, which is available at a fraction of the cost of
higher purity material.
The substrate (2) can be quite thin. One type of suitable
substrate is polycrystalline Ni foil (25 µp?, 99.5%,) or also
polycrystalline Ni foil (25 \im, 99%,).
Without wishing to be bound by theory, it is believed that Ni acts
as a catalyst for the breakdown of hydrocarbon species into
smaller building units essential for the synthesis of graphene.
Other transition metals can be used with minor modification. For
instance, while Nickel is a useful substrate under ambient
atmospheric conditions, Copper can be used as a substrate for the
growth of graphene by evacuating any ambient air within the tube
at the start of the process. The remainder of the process is
otherwise the same. However, regardless of the substrate, the
methods of the present invention avoid the use of expensive
compressed gases as required in prior art methods. The carbon
source can be any source of materia! thai provides volatile carbon
at temperatures between 2GG-650°C at ambient pressures. For
instance, animal or vegetable fat in unprocessed form have both
been found to be useful.
One particularly useful source of carbon is raw soybean oil, which
is a triglyceride of formula C8H3606. More abundant biomass and
industrial by-products, for example, cei!u!osic materials, may be
used. The present inventors have established that there is no need
to use highly purified material as the carbon source. A minima!
quantity is used, for example Q.1 - 0.25 mL of soybean oil
provided sufficient carbon to coat an area of 10cm<2>of
metal substrate. If a thicker graphene film is required, G.25-0.30
mL (or g) of carbon source per 10cm<2>of metal substrate can
be used.
If too much carbon source materia! is used, the graphene film will
be thick, for instance, graphene films of about 5-10 layers are
obtained. Similarly, if significantly lower amounts of carbon
source than is optimal, then graphene formation will not occur and
amorphous carbon obtained. The furnace temperature is then raised
to around 800°C over a period of 20-30 minutes. A typical ramping
rate is from 25-35°C/min). During the ramping stage (~300°C-350°C)
the precursor is vaporized and the long carbon chains in the
soybean oil begin to be broken down into gaseous carbon building
units via thermal dissociation. Those skilled in the art will
appreciate that the precise dissociation temperature will differ
based upon the chemical and physical properties of carbon source
precursor material. Simultaneously, gaseous carbon building units
diffuse throughout the tube and towards the Ni foii growth
substrate. As the temperature in the furnace gradually increases
to 80QX, the carbon precursor is further broken down into simpler
carbon units for graphene generation on the surface of metal
substrate. In addition, as the temperature rises the carbon
solubility in Ni increases and the carbon building units begin to
dissolve into the Ni bulk. From 500°C a graphitization process
takes place where carbon atoms are starting to arrange themselves
in sp<2>configuration. From 500°C to 800°C graphene lattice
is shaped.
Graphene formation is observed to take place from 650°C, although
the best quality graphene (in terms of low defects) is obtained
from about 8Q0°C.
Graphene also grows at higher temperatures Whilst it is
advantageous to carry out the growth at the lowest possible
temperatures, it has been found that conducting the growth process
at 900°C, rather than 8Q0°C results in a thicker graphene layer
which is less hydrophobic in character. The contact angle at 900°C
is 140.2° compared to 96.1 ° for graphene films grown at 800°C, so
by varying the growth parameters the contact angle can be varied
significantly by an amount up to at least 40° to 45°. Without
wishing to be bound by theory, it is possible that temperature
could be dependent on the chemical makeup of the carbon precursor
chemistry, for instance a simpler carbon chain could enable
graphene production at a lower temperature.
Once the desired temperature is reached, the furnace is held at
that temperature, for example, 800°C (for 1 G~15min for 99.5%
purity Ni foil) to enable growth. Graphene grains enlarge during
the annealing process. The annealing time can be reduced by using
lower purity films. For instance, the annealing time can be
reduced to around 3 minutes if 99% purity Mi foil is used.
Following the completion of the growth period, the tube is
immediately removed from the heating zone to enable the controlled
cooling to take place.
Control of the cooling rate is a key step in the process of the
present invention. During the cooling stage, carbon segregates
from the metal bulk, resulting in the formation of graphene on the
metal surface, if cooling is controlled to be too slow,
graphite-like thick graphene films (>10 graphene layers) are
obtained. The presence of ambient air during the cooling process
is also thought to etch the graphene film and helps the formation
of high quality graphene film of typically 1-3 layers. The use of
a Nickel substrate does not appear to be adversely affected by the
presence of air, however, Copper substrates provide more growth of
graphene domains in the absence of any gas, i.e. under vacuum.
Substrates that are more susceptible to competing oxidation
reactions would advantageously be reacted under conditions
requiring the additional evacuation step. Both the amount of
precursor and the cooling rate independently affect the resultant
film, if good quality films of greater thickness than 2-3 atomic
layers, for instance, 10 or more, 40 or more, or even 100 or more
layers are required, it is advantageous to increase the amount of
precursor whist at the same time dropping the cooling rate. The
thickness of the films is measured by the optical transmission,
whereas quality is determined by \Gand /3/476ratios which are well
established Raman spectroscopic parameters for determining
graphene quality
At a fast cooling rate, typically 25°C/min and optimal precursor
amount (typically ~ 0.15 mL per 10cm<2>of substrate), thin
films with 2-3 layers of graphene with good qualityGup to 0.4 and
/2D/Gup to 2.5) are observed. Reducing the cooling rate to 18C/min
and increasing the precursor amount (typically -0.2 mL per
10cm<2>of substrate), thicker films with around 20 layers of
graphene of good quality (/D//Gup to 0.4 and /2D//Gup to 1.8) are
observed. Reducing the cooling rate even further, to around
10C/min and significantly increasing the precursor amount
(typically - 0.25 mL per 10cm<2>of substrate), gives thick
films of graphene with around 40 layers with a very high quality
(/r//Gup to 0.4 and !2D/'IGup to 0.4) Figure 2 shows the proposed
mechanism for graphene growth, as well as the temperature profile
of a preferred embodiment of the process. It was observed thai the
carbon precursor content and sample cooling rate significantly
influenced the quality of the subsequently produced graphene
films. In an ambient-air environment, an insufficient amount of
precursor material leads to the presence of O and N species in the
subsequent graphene, in the form of C-O or C-N amorphous carbons.
This may suggest the critical role of the thermally dissociated
precursor (soybean oil in the invention as exemplified) in
consuming the O and N species present in the ambient-air
environment.
It is thought that an excessive amount of precursor material may
lead to an oversaturation of deposited carbon in the metal
substrate, which can lead to crystallization of carbon on the
metal surface before the cooling process. This may explain the
resulting formation of undesirably thick graphene sheets observed
when excessive carbon source precursor is used.
A slow sample cooling rate may, on the other hand, promote
excessive carbon segregation from the metal bulk during the
cooling process, and may account the observed formation of
graphite.
A number of substrates were tested under the same experimental
conditions used for the nickel substrate. Even when these
substrates did not exhibit graphene formation, they did show that
a graphitization process (sp<0>to sp<2>transition of
the carbon) had occurred. The graphitization process precedes the
possible arrangement of carbon atoms into graphene.
This observation may suggest the key role of the properties of the
substrate (such as carbon solubility and catalytic effect) in
graphene formation. The process for ambient-air growth of graphene
of the present invention has been successfully applied using other
types of carbon-containing precursors, including triglycerides
such as butter, which produced high quality singie-tri layer
graphene films. It is believed that the method of the present
invention is versatile and may be tailored to transform other
renewable and redundant biomasses into graphene films. it is
important to note that throughout the process, the pressure in the
container is maintained at atmospheric pressure for the Nickel
substrate. Throughout the entire growth process, no additional
gases were introduced into the quartz tube. After cooling, the
substrate (2) was removed and the graphene (6) grown thereon was
analysed.
The following table shows a comparison between the methods of the
present invention and the leading articles relating to graphene
synthesis.
Table 1 Comparison of Synthesis Methods,
1. X. Li, W. Cai, J. An, S. Kim, J. Nah. D. Yang, R. Finer. A.
Velamakanni. I. Jung, E. Tutuc, S. K. Banerjee. L. Colombo and R.
S. Ruoff, Science, 2009, 324, 1312-1314.
2. K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, K. S. Kim,
J.-H. Ahn. P. Kim, J.-Y. Choi and B. H. Hong, Nature, 2009, 457,
706-710.
3. G. Ruan, Z. Sun, Z. Peng and J. . Tour, ACS Nano, 2011 , 5,
7601 -7607.
4. S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J.
Baiakrishnan, T. Lei, H. Ri Kim, Y. i. Song, Y.-J. Kim, K. S. Kim,
B. Ozvilmaz, J.-H Ahn, B H. Hong and S lijima, Nat Nano, 20 0, 5,
574-578.
The present method thus provides many advantages over other
methods, including the ability to use a renewable low quality
biomass, air at atmospheric pressure and lower temperatures.
The use of lower temperatures and shorter processing times is also
energetically favourable. As a result, the present invention can
produce graphene for around 14% of the cost of conventional
methods.
The quality of the graphene produced by the methods of the present
invention was compared to graphene available from commercial
manufacturers. The results are presented in table 2 below. The
details were sourced from the Graphene Market Report 2015,
FutureMarketslnc, and websites of the respective companies. Ail
the commercial graphene films described in table 2 were prepared
using conventional technology under purified gas atmospheres.
Table 2 Comparison of Graphene Properties.
Company Transmittance (%) Film Resistance (Q/sq)
Present invention 94.3 330
2D Carbon Tech 85 200-400
BGT Materials 95 800
Graphenea 97 450
Graphene Square 97 250-400
Wuxi Graphene Film 97 600 It cars be seen thai the present
invention produces graphene with a very high transmittance, making
it very suitable for optical uses. In addition, the resistance of
the graphene produced by the present methods is very low, meaning
the graphene is relatively defect-free and highly suitable for
electronic uses.
The present invention allows for the synthesis of high quality
graphene films to take place in an ambient-air environment via
thermal chemical vapour deposition. The absence of a vacuum
chamber means that the present process can be highly scalable.
Ambient-air synthesis according to the present invention
facilitates a streamlined integration into the large-scale
graphene production infrastructure such as roli-to-roii or batch
processing required for industrial production.
The present invention allows for thermal-based synthesis in the
absence of any purified compressed feedstock gases (e.g., methane,
hydrogen, argon, nitrogen, etc.), which are cosily and/or highly
explosive. The synthesis technique of the present invention does
not require any purified feedstock gases, and instead, can utilize
far cheaper carbon source material such as a renewable biomass as
the precursor for the synthesis of graphene films. Notably, this
enables the process of the present invention to be technologically
sustainable, and also significantly cheaper and safer than
presently available methods. The methods of the present invention
significantly reduce the processing times compared to techniques
presently described in the literature and those adopted by
industry. The present process involves only a single step, and
requires significantly less processing time. This is unlike
existing methods for the synthesis of graphene films, which
involve sample pre-treatments in purified gas environments at high
temperatures and much longer processing time.
The present methods are thus safe, environmentally-friendly, and
resource-efficient technique for graphene synthesis.
EXAMPLES
EXAMPLE 1. OPTICALLY TRANSPARENT CONDUCTIVE FILM
In one embodiment, the controlled growth conditions of the present
invention could be used to produce a thin (optically transparent)
graphene film useful in a sensing device. Films of materials which
are optically transparent in a practical sense are those which
exhibit optical transmissions of 80 - 100% in the visible
wavelength range, similar to the transparencies achieved by indium
tin oxide (ITO) films Growth of Graphene
The growth oi graphene was carried out in a thermal CVD furnace
(OTF-12Q0X-UL, MTI Corp). A quartz tube was used. Polycrystalline
Ni foils (25 µ??, 99.5% or 99%, Alfa Aesar) were used as the
growth substrate.
Two alumina crucibles were loaded into a quartz tube. One crucible
contained the carbon source, which was 0.1 -0.25 mi_ of soybean
oil. The other crucible held a square (10cm<2>) of the Ni
foil growth substrate. These two crucibles were placed close
proximity within the quartz tube. The tube was positioned so that
both crucibles were within the heating zone of the furnace. The
open ends of the quartz tube were then sealed.
The furnace temperature was raised to 8GG C (30 C/min) followed by
maintaining the temperature for 15 mins for 99.5% purity Ni foil
and 3mins for 99% purity Ni foil at 800<°>C to form a
graphene lattice. Following lattice formation, the growth
substrate was immediately removed from the heating zone to enable
cooling at a controlled rate (50-100<°>C/min) to allow
segregation of the graphene lattice from the metal substrate to
form a deposited graphene.
The pressure in the tube was maintained at ambient pressure.
Throughout the entire growth process, no additional gases were
introduced into the quartz tube.
Once cooled to ambient temperature, the substrate was removed from
the tube and the as-grown graphene film was analysed using
conventional techniques, as described below. The film resistance
was very low, 330O ohms/sq using four probe sheet resistance
measurements and the visible spectrum transmittance was 94.3%,
Also Raman spectra indicate that graphene is formed with a
relatively low proportion of defects and being very thin (three or
less films). These characterisation suggest that this graphene
obtain from this process is high quality.
The functional properties of the graphene of the present invention
are also very advantageous, as shown by its ability to act as a
genosensor, which is detailed below.
A poly (methyl methacryiate) (P A)-assisted transfer of graphene
was used. 48 mg/mL of PIV1 A (Mw998,000) was spin-coated on the
as-grown graphene on Ni foil (3000 rpm for 1 min). The sample was
then dried in open air for 12 hours. Subsequently, the underlying
Ni foil was dissolved in 1 M FeCI3in 30 minutes. The PMMA/graphene
film then floated to the surface. This was washed several times
with deionised water. Next, the PMMA/graphene was lifted off from
the deionised water bath and transferred onto a glass substrate.
The PMMA was then dissolved with acetone, and the sample was
repeatedly washed with deionised water. The graphene isolated on
glass was then used for subsequent microscopy and electrical
characterizations. Microscopy and microanalysis
Field-emission scanning electron microscopic (FE-SEM) images were
obtained by Zeiss Auriga microscope operated at 5 keV electron
beam energy with an InLens secondary electron detector. Raman
spectroscopy was performed using a Renishaw inVia spectrometer
with a laser excitation at. 514 nm (Ar laser) and a probing spot
size of -1 µp?<2>. X-ray photoeiectron spectroscopy (XPS)
spectra were recorded by Specs SAGE 150 spectroscope with the IV!g
Ka excitation at 1253.6 eV. Both survey and narrow scans of C 1 s
and O 1 s were conducted. Transmission electron microscopic (TEM)
images were obtained by the JEOL 220QFS TE microscope operated at
200 kv.
Optical characterizations Optical images were obtained by an
Olympus BX51 optical microscope. Transmittance measurements were
obtained by a Varian Cary 5000 UV-Vis spectrophotometer. A
graphene area of 4 cm<2>was used, and optical spectra were
recorded in the wavelengths of 300 - 800 nm.
Electrical four-probe measurements
Silver paint was applied to the graphene transferred onto glass. A
graphene area of 1 cm<2>was used. Four-point probe
measurements were conducted at room temperature.
The topographic characterizations in Figure 3 demonstrate the
uniformity and coverage of the as- grown graphene. A continuous,
smooth, and large-area graphene film with good homogeneity was
observed to cover the entire surface of the Ni foil. The optical
micrograph in Figure 3(a) indicates an average graphene grain size
of 30-200 µ??. The large-area of transparent graphene film is
evidenced by the low-magnification TEM image (Figure 3(b)) and the
respective electron diffraction pattern (inset). Correspondingly,
a Fast Fourier Transform (FFT) spectra in Figure 3(c) further
supports the presence of graphene lattices, and TEM
characterizations in Figure 3(e)-(g) indicate the presence of 1 -3
graphene layers in the graphene film. Furthermore, the graphene
films are observed to be continuous across the Ni grain
boundaries, as evidenced by close-up SEM images which show the
wrinkled surface of the graphene films. Such morphological
features stem from the difference in coefficient of thermal
expansion between graphene and Ni, and indicate continuity of the
graphene film as these wrinkles span the grain boundary.
The surface of the as-grown graphene was characterized by Raman
and X-ray photoeiectron spectroscopy (XPS). The Raman spectra in
Figure 4 illustrate the transformation of the precursor at
different stages of annealing temperatures. In the Raman spectra,
three distinct peaks are present, namely, the characteristic
disorder peak (D-band) at 1350 cm ' , the graphitic peak (G- band)
at 1580 cm<"1>, and the second-order 2D-band at 2670
cm<">' . The G-band arises from the in- plane vibrational
E2gmode of the sp<2>-hybridized carbon, the D-band is
attributed to the finite crystallite size effect and various
defects induced in the sp<2>carbon materials, and the
2D-band is a second-order Raman spectral feature due to the
three-dimensional inter-planar stacking of hexagonal carbon
networks. At low temperatures of 500°C, there is insufficient
energy provided to breakdown the soybean oil precursor into
smaller hydrocarbon building units for graphene (Figure 4(a)). The
transition of graphene quality with increasing temperatures is
evident (Figure 4(b)), and large area low-defect graphene sheets
are produced as shown with an annealing temperature of
800<°>C (Figure 4(c)). A t 800°C, the ratios of Raman
!ZD/!G-(0.9-2.2) and iG-(0.1 -0.25) indicate the presence of 1 -3
graphene layers with low defects and a dominant
sp<2>graphitic structure. These results show that
temperature is a key parameter which controls the nucleation and
growth of graphene from the carbon source precursor.
The XPS characterizations in Figures 5(a)-(b) show a single and
narrow C 1 s peak at the binding energy (BE) of 284.5eV, which
indicates the nanostructure is made of mostly carbon atoms. The
survey scan of Figure 5(a) shows a small peak intensity of the C 1
s in relation to the Ni LMM, and suggests the presence of an
atomically thin graphene film such that the electron beam may
penetrate through to the underlying Ni foil growth substrate. The
corresponding C 1 s narrow scan in Figure 5(b) may be fitted by
three peaks corresponding to the carbon sp'<'>(BE - 284,5
eV), sp<3>(BE - 285.4 eV), and the oxygen-attached carbons
(BE - 286 eV). The graphene of the present invention demonstrates
high ratios of sfflsp<3>carbons and C/O. This indicates the
graphene lattice is of high structural quality, and its growth was
not affected by the N and O species present in the ambient-air
environment. The optical transmittance of the transferred graphene
film is shown in Figure 5(c). A peak transmittance of 94,3% at.
550 nm is observed, and suggests a film thickness averaging 2
graphene layers. Figure 5(d) illustrates the conductivity
measurements of our graphene, where a sheet resistance 330 Ohm/sq
is reported. These results suggest thai the graphene of the
present invention possesses a low level of defects and a highly
crystalline lattice, leading to its excellent optoelectrical
properties which are comparable to if not better than those
obtained by conventional CVD methods. Thus, our as-grown graphene
films are highly promising nanomaterials for integration in a host
of electronic devices.
The method of the present invention can be quite readily scaled to
larger area sample by simple control of precursor flow to enable
high quality graphene films of different dimensions to be
produced, in the case of the present invention, the process for
preparing a 4cm x 2.5cm strip was readily scaled to a 1 Gem x
2.5cm strip with no complications or drop in quality of the
graphene film.
Biosensor device
The present invention provides access to large-area, uniform, and
continuous graphene films. One such potential application for this
material is in the area of high-throughput electrochemical
biosensirsg devices. Electrochemical sensing methods for minute
amounts of nucleic acid samples offer attractive opportunities for
decentralized genetic testing and a plethora of other preventative
health technologies, which require portable, cost-effective, and
low-power readout devices, Neurodegenerative diseases such as
Alzheimer's disease are candidates for such testing. These
diseases are becoming more prevalent with the ageing population.
At present, over 40 million individuals worldwide are affected by
Alzheimer's disease and this may treble by 2050. Importantly,
Alzheimer's disease may be best managed with early intervention
therapies provided it is diagnosed as early as possible. Recently,
post-transcriptional epigenetic regulations of gene expressions
have been found to provide highly-valuable serum-based nucleic
acid biomarkers which may be utilized to enable early diagnostic
strategies for the disease. Graphene, particularly high quality
graphene in large sheets as produced by the present invention, can
be used as biosensing electrode to detect and provide early
diagnosis for Alzheimer's disease. The graphene produced by the
method of the present invention was used to prepare an
electrochemical genosensor for Alzheimer's disease to demonstrate
the excellent functionality of the graphene films of the present
invention. The general configuration of the biosensor is set out
in figure 6. Biosensor device assembiy
The as-grown graphene on Ni foil was treated with a
low-temperature 02plasma (100W, 7 seconds) to introduce carboxyl
functional groups on its surface. The sample was placed flat and 2
cm below the plasma generation zone. The size of each sensing
substrate was 2 cm x 1 cm. Subsequently, the plasma-activated
graphene was treated with 0.5 M A/-(3-Dimethyiaminopropyi)-
W-ethylcarbodiimide hydrochloride (EDC) and 0.03 M
N-hydroxysulfosuccinimide (NHS) in phosphate buffered saline (PBS,
pH = 7,) for 15 mins. This enabled the formation of active ester
intermediates via carbodismide chemistry. Next, the surface of
graphene was washed several times with phosphate buffered saline
(PBS, pH 7, Sigma Aldrich) and Dl water to remove excess EDC/NHS.
Next, NH2-conjugated miRNAs (probe sequence:
5'-NH2-GGTGGAGGGGACGTTTGCAGGT-3') were diluted in PBS to 0.2 µ?,
and 50 µ?_ was pipetted onto the EDC/NHS-ireated surface. This was
left to incubate overnight in a wet environment and at room
temperature. Next, the sensing surface was washed with Q.05%
sodium clodecyl sufonate (SDS) in 0.Q4 M hydroxylamine solution to
deactivate the remaining carboxyl functional groups and to remove
non-specificaiiy bound probe miRNAs. Then, 0.01 M Polyethylene
glycol (PEG) was loaded on the sensing surface to block the
exposed areas of graphene to reduce further non-specific binding.
Next., the ibiomarker) miRNA sequence (target sequence:
5'-CCACCUCCCCUGCAAACGUCCA-3') was dissolved in human serum (Human
Plasma AB) to obtain dynamic concentrations of 1 nM - 0.1 pM, that
were pipetted onto the sensing surface. This was ieft to incubate
at 45<°>C for 20 mins to induce hybridization between the
complementary probe and target sequences.
Finally, a washing step with PBS/DI water was employed to remove
remaining non-specifically bounded target miRNAs. To demonstrate
sensing specificity, a similar protocol was adopted by replacing
the target sequence with a single-base mismatched miRNA sequence
(non- complementary sequence: S'-CCGCCUCCCCUGCAAACGUCCA-S'). This
fully assembled device was then utilized in a three-electrode
electrochemical ceil for biosensing measurements. Bsoserssirsg
measurements
The electrochemical measurements were conducted in 10 m FeCN6in
0.1 M Na2S04at room temperature. A three-electrode ceil
configuration was employed. The three electrode cell used the
as-grown graphene on Ni as the working electrode, a Pt wire as the
counter electrode, and an Ag/AgCI reference electrode ( etroOhm).
The EIS measurements were conducted in the frequency range from
500 kHz - 1 kHz, using a BioLogic VSP 300 potentiostat/gavanostat
instrument. The charge-transfer resistance Rdof the sensor
following incubation with the target miRNA was expressed as a
percentage of Rdin the reference (blank) case, which was incubated
in the human serum medium in the absence of target miRNAs.
The impedance spectra in Figures 6(b)-(c) demonstrate the change
in electrode-electrolyte resistance (Rc<) at the biosensing
electrode interface due to selective immobilization of target
anaiyte. Notably, the device of the present invention exhibits a
bio-detection sensitivity of Q.1 pM, with device selectivity down
to single-base mismatched sequences, and a large dynamic sensing
range (1 nm - 0.1 pM). We define ARctby (Rct- RQ)IR0, where R0is
the charge-transfer resistance of the reference sample. This
increase in ARCTmay be attributed to a retarded charge transport
towards the graphene surface, either through spatial blocking or
electrical repulsion, in particular, the hybridization between
complementary genomic sequences induces a build-up of negative
surface charge, and the repulsion of negatively charged
ferricyanide ions, which leads to a rise in Rct In addition, the
device shows negligible response to the non-complementary miRNA.
This performance is comparable to commercial ELISAs, and other
graphene-based genosensors reported in the recent literature.
importantly, the integration of the graphene of the present
invention as a biosensing platform enables a highly specific
detection of miRNA biomarkers at physioiogicaiiy-reievant
concentrations. Thus, this may enable an early diagnosis of
Alzheimer's disease, which requires the quantification of multiple
miRNA biomarkers in complex biological environments.
Electrochemical graphene- based biosensors are advantageous over
ELISAs and other biosensing techniques, as these either require
resource-consuming (energy, time) fabrication for bio-defection
(e.g., fluorescent tags, lithography, electrode patterning, etc.),
large sample amounts (of miRNA, i.e. extracted blood from
patient), or multi-staged amplification bioassays to quantify the
concentration of target analyte. Currently, the one of the main
drawbacks with integrating graphene for medical sensing is its
cost for production functional graphene. However, the method of
the present invention demonstrate that high-quality functional
graphene films can be grown using a simple, cheap and industrially
viable technique, and that the graphene can be successfully
integrated as a genosensor for Alzheimer's disease. The
high-throughput electrochemical performance of our graphene
genosensor provides a practical demonstration of the high quality
of graphene produced by the present invention.
Electrochemical graphene-based genosensors may be tailored to
different types of other diseases, epidemics, including also
point-of-care diagnostics for infectious diseases.
EXAMPLE 2. CONTROLLED THICKNESS GRAPHENES
The growth of graphene was carried out as described for example 1
, with modification to the amount of graphene and the cooling
rate. A quartz tube was used. Polycrystaiiine Ni foils (25 µ??,
99.5% or 99%,) were used as the growth substrate.
Two alumina crucibles were loaded into a quartz tube. One crucible
contained the carbon source, the other crucible held the Ni foil
growth substrate. These two crucibles were placed in close
proximity inside the quartz tube. The tube was positioned so that
both crucibles were within the heating zone of the furnace. The
open ends of the quartz tube were then sealed.
The furnace temperature was raised to 800 C (30<°>C/min)
followed by maintaining the temperature for 15 mins to allow
graphene lattice formation on 99.5% purity Ni foil and 3mins for
99% purity Ni foil at 800 C.
After the growth step, the growth substrate was immediately
removed from the heating zone and cooled at the controlled rate.
The pressure in the tube was maintained at ambient pressure.
Throughout the entire growth process, no additional gases were
introduced into the quartz tube. Once cooled to ambient,
temperature, the substrate was removed from the tube and analysed.
Cooling rate / Precursor amount / Resultant film
25°C/min 0.15 mL per 10cm<2>of substrate Approximately 2-3
layers
18°C/min 0.2 mL per i 0cm<z>of substrate Approximately 20
layers
10°C/min 0.25mL per l Ocm"<2>of substrate Approximately 40
layers
With increased amount of precursor material and decreased cooling
rate of sample, graphene samples are observed to increase in
thickness and reduce in defect content. This transition is
illustrated in Figure 7. in Fig. 7 (a), when a fast cooling rate
and typically -0.15 ml_ of precursor per 10cm<2>of substrate
is used, the graphene films demonstrate an optical transmission of
-94.3%, yiG- 0.1 - 0.33 and l?D/!G- 0.85 - 2.5. These
characterizations are consistent with good quality graphene films
of 2 - 3 layers. in Fig. 7 (b), at a reduced cooling rate and with
typically -0.2 mL of precursor per 10cm<2>of substrate being
used, optical transmission drops to -58.9%, yiG-0.1 - 0.33 and
l2D/lG- 0.4 - 1 .6. These properties of graphene film are
consistent with a film of -20 layers of graphene.
Finally, in Fig. 7 (c), at the slowest cooling rate and with
typically -0.25 mL of precursor per 10cm<2>of substrate
being used, optical transmission is -23.9%, lD/lGis - 0 - 0.3, and
/2D/Gis mostly - 0.4. This indicates a film with - 40 layers with
low/no defects.
Without wishing to be bound by theory, a slower cooling rate is
believed to provides more time for carbon to segregate to the
surface of the Ni foil growth substrate, and an increase in
precursor amount directly increases the carbon content within the
processing chamber, and therefore, these modifications in growth
conditions result in an increased in thickness of the graphene
films.
EXAMPLE 3. HYDROPHOBICITY/HYDROPHILICITY CONTROL
Graphene was grown on a Nickel substrate in accordance with the
procedures described above. The growth temperature was raised to
900°C and maintained for a time period between 3-1 5 minutes. The
results are shown in Fig 8 (a) - (b), which demonstrate Raman
ratios of yiG- 0.1 - 0.3 and l2D/lG- 0.1 - 0.7, which indicate the
presence of about 40 layers of graphene films with low/no defects.
Again, without wishing to be bound by theory, it is believe that
this may be attributed to an overall longer annealing time and
faster carbon diffusion rate at. a higher temperature, that may
allow for more carbon to segregate from the Ni foil growth
substrate, leading to overall thicker films.
Contact angle measurements for graphene films grown at 9Q0°C had a
contact angle of 140.2° compared to 96.1 ° for those grown at
8G0°C. The films prepared at 900°C showed a decreased in
hydrophobic character compared to those grown at 800°C,
EXAMPLE 4. COPPER SUBSTRATE
The growth of graphene was carried out in a thermal CVD furnace
(OTF-12QQX-UL, MTi Corp). A quartz tube was used, A copper foil
(0.025mm, 99.8%) was used as the growth substrate. Two alumina
crucibles were loaded into a quartz tube. One crucible contained
the carbon source, which was 0.1 -0.25 mL of soybean oil. The
other crucible held a square (10cm<2>) of the copper foil
growth substrate. These two crucibles were placed close proximity
within the quartz tube. The tube was positioned so that both
crucibles were within the heating zone of the furnace. The open
ends of the quartz tube were then sealed.
The tube was then evacuated to a vacuum of less than 1 mm Hg. The
furnace temperature was raised to 800<°>C (30<°>C/min)
followed by an annealing for (3 minutes) 800 C. After the
annealing step, the growth substrate was immediately removed from
the heating zone to enable a controlled cooling (rate of
50°C/min).
The vacuum in the tube was maintained until the system had
returned to room temperature. Throughout the entire growth
process, no additional gases were introduced into the quartz tube.
Once cooled to ambient temperature, the substrate was removed from
the tube and the as-grown graphene film was analysed using
conventional techniques. Figure 9 shows Raman spectral
characterization of graphene grown on copper foil, made possible
by modification of growth conditions in our process. The as-grown
graphene indicated the presence of small graphitic islands, spread
over the surface of the copper foil growth substrate. These
graphitic islands indicated film quality of /D//G- 0.6 and were of
multi-layer thickness (/2D//G~0.5).
Corsdussors
The present invention provides a method for synthesis of
high-quality, large-area, uniform, and continuous graphene films
of a controlled thickness and hydrophobicity, in an atmospheric
pressure, compressed gas-free and ambient-air environment. The
method is highly scalable, resource-efficient, simple,
single-stepped and environmentally-friendly, and offers numerous
advantages over conventional thermal CVD techniques for graphene
synthesis, which critically rely on numerous resource-consuming
procedures and expensive consumables. These methods offer numerous
advantages and opportunities for future streamlined integration in
large-scale production infrastructures. The functionality of the
as-grown graphene films has been demonstrated by its direct
integration into a functioning high performance electrochemical
genosensor for Alzheimer's disease.